Physical Sciences and Mathematics Commons^™

Machine Learning With Topological Data Analysis, Ephraim Robert Love

Doctoral Dissertations

Topological Data Analysis (TDA) is a relatively new focus in the fields of statistics and machine learning. Methods of exploiting the geometry of data, such as clustering, have proven theoretically and empirically invaluable. TDA provides a general framework within which to study topological invariants (shapes) of data, which are more robust to noise and can recover information on higher dimensional features than immediately apparent in the data. A common tool for conducting TDA is persistence homology, which measures the significance of these invariants. Persistence homology has prominent realizations in methods of data visualization, statistics and machine learning. Extending ML with …

A Framework To Detect The Susceptibility Of Employees To Social Engineering Attacks, Hashim H. Alneami

Doctoral Dissertations and Master's Theses

Social engineering attacks (SE-attacks) in enterprises are hastily growing and are becoming increasingly sophisticated. Generally, SE-attacks involve the psychological manipulation of employees into revealing confidential and valuable company data to cybercriminals. The ramifications could bring devastating financial and irreparable reputation loss to the companies. Because SE-attacks involve a human element, preventing these attacks can be tricky and challenging and has become a topic of interest for many researchers and security experts. While methods exist for detecting SE-attacks, our literature review of existing methods identified many crucial factors such as the national cultural, organizational, and personality traits of employees that enable …

Design And Control Of A Peristaltic Pump To Simulate Left Atrial Pressure In A Conductive Silicone Model, Jeremy Collins

Mechanical Engineering Undergraduate Honors Theses

According to the CDC, atrial fibrillation is responsible for more than 454,000 hospitalizations and approximately 158,000 deaths per year. A common treatment for atrial fibrillation is catheter ablation, a process in which a long flexible tube is guided through the femoral artery and to the source of arrhythmia in the heart, where it measures the electrical potential at various locations and converts problematic heart tissue to scar tissue via ablation. This paper details the design and control of a low-cost ($400) peristaltic pump system using repetitive control to replicate blood pressure in the left atrium in a conductive silicone model …

Data Forgery Detection In Automatic Generation Control: Exploration Of Automated Parameter Generation And Low-Rate Attacks, Yatish R. Dubasi

Computer Science and Computer Engineering Undergraduate Honors Theses

Automatic Generation Control (AGC) is a key control system utilized in electric power systems. AGC uses frequency and tie-line power flow measurements to determine the Area Control Error (ACE). ACE is then used by the AGC to adjust power generation and maintain an acceptable power system frequency. Attackers might inject false frequency and/or tie-line power flow measurements to mislead AGC into falsely adjusting power generation, which can harm power system operations. Various data forgery detection models are studied in this thesis. First, to make the use of predictive detection models easier for users, we propose a method for automated generation …

Wound Image Classification Using Deep Convolutional Neural Networks, Behrouz Rostami

Theses and Dissertations

Artificial Intelligence (AI) includes subfields like Machine Learning (ML) and DeepLearning (DL) and discusses intelligent systems that mimic human behaviors. ML has been used in a wide range of fields. Particularly in the healthcare domain, medical images often need to be carefully processed via such operations as classification and segmentation. Unlike traditional ML methods, DL algorithms are based on deep neural networks that are trained on a large amount of labeled data to extract features without human intervention. DL algorithms have become popular and powerful in classifying and segmenting medical images in recent years. In this thesis, we shall study …

Cortical Dynamics Of Language, Kiefer Forseth

Dissertations & Theses (Open Access)

The human capability for fluent speech profoundly directs inter-personal communication and, by extension, self-expression. Language is lost in millions of people each year due to trauma, stroke, neurodegeneration, and neoplasms with devastating impact to social interaction and quality of life. The following investigations were designed to elucidate the neurobiological foundation of speech production, building towards a universal cognitive model of language in the brain. Understanding the dynamical mechanisms supporting cortical network behavior will significantly advance the understanding of how both focal and disconnection injuries yield neurological deficits, informing the development of therapeutic approaches.

Direct Drive Solar Panel Control Circuit, Marcorios Bekheit

Physics

A control circuit is built for insulated solar electric cookers (ISEC). Power delivery and temperature safety are the focus. Using a maximum power point tracking (MPPT) algorithm, Arduino Nano, voltage and current sensors, and a buck converter, the solar panel’s output power was maximized for a direct load heat resistor with 3.5Ω for a range of solar intensities. Using a resistance temperature detector, a temperature sensor is built for safety shutoff.

Quantum Dynamical Phenomena In Non-Hermitian And Magnomechanical Systems, Saeid Vashahri Ghamsari

Graduate Theses and Dissertations

In this dissertation, we have investigated quantum dynamics via three case studies. First, we studied a system of two coupled waveguides respectively carrying optical damping and optical gain in addition to squeezing elements in one or both waveguides. Such a system is expected to generate highly entangled light fields in the two waveguides. We, however, show that the degree of the created entanglement is significantly affected by the quantum noises associated with the amplification and dissipation. Because of the noise effect, one can only have nonzero entanglement for a limited time interval. Second, we generalized the first project by considering …

Achieving Differential Privacy And Fairness In Machine Learning, Depeng Xu

Graduate Theses and Dissertations

Machine learning algorithms are used to make decisions in various applications, such as recruiting, lending and policing. These algorithms rely on large amounts of sensitive individual information to work properly. Hence, there are sociological concerns about machine learning algorithms on matters like privacy and fairness. Currently, many studies only focus on protecting individual privacy or ensuring fairness of algorithms separately without taking consideration of their connection. However, there are new challenges arising in privacy preserving and fairness-aware machine learning. On one hand, there is fairness within the private model, i.e., how to meet both privacy and fairness requirements simultaneously in …

Memory Module Design For High-Temperature Applications In Sic Cmos Technology, Affan Abbasi

Graduate Theses and Dissertations

The wide bandgap (WBG) characteristics of SiC play a significant and disruptive role in the power electronics industry. The same characteristics make this material a viable choice for high-temperature electronics systems. Leveraging the high-temperature capability of SiC is crucial to automotive, space exploration, aerospace, deep well drilling, and gas turbines. A significant issue with the high-temperature operation is the exponential increase in leakage current. The lower intrinsic carrier concentration of SiC (10-9 cm-3) compared to Si (1010 cm-3) leads to lower leakage over temperature. Several researchers have demonstrated analog and digital circuits designed in SiC. However, a memory module is …

Analog Spiking Neural Network Implementing Spike Timing-Dependent Plasticity On 65 Nm Cmos, Luke Vincent

Graduate Theses and Dissertations

Machine learning is a rapidly accelerating tool and technology used for countless applications in the modern world. There are many digital algorithms to deploy a machine learning program, but the most advanced and well-known algorithm is the artificial neural network (ANN). While ANNs demonstrate impressive reinforcement learning behaviors, they require large power consumption to operate. Therefore, an analog spiking neural network (SNN) implementing spike timing-dependent plasticity is proposed, developed, and tested to demonstrate equivalent learning abilities with fractional power consumption compared to its digital adversary.

Thruster Communication For Subsurface Environments; Turning Waste Noise Into Useful Data, Stephen Cronin

Doctoral Dissertations and Master's Theses

Acoustic communication serves as one of the primary means of wirelessly communicating underwater. Whereas much of the developments in the field of wireless communication have focused on radio frequency technology, water highly absorbs radio waves rendering the link not feasible for most all subsurface operations. While acoustic links have enabled new capabilities for systems operating in this challenging environment, it has yet to reach the commodity availability of radio systems, meaning that an entire class of small, low-cost systems have been unable to make use of these links. The systems in question are primarily autonomous underwater vehicles (AUVs), as they …

Dynamic Analysis Of A Microgrid Powered With An Inverter And Machine-Based Distributed Resources, Brandon William Blackstone

UNLV Theses, Dissertations, Professional Papers, and Capstones

The proliferation of renewable distributed energy resources, particularly photovoltaic (PV) power systems, and the increasing need for a reliable power supply has led to the concept of microgrids, a mini-grid that consists of locally connected power generation units and needs, able to operate connected or disconnected from the utility grid, using controlled and coordinated methods to provide for the users of the microgrid the best possible conditions for their needs. The main technical issues facing microgrids include some of the following, seamless transition from stand-alone to utility grid connected operation, how to preserve frequency and voltage stability, and provide the …

App Development For Wearable Sensors, Connor Pittman, Christopher Chappie, Patrick A. Tetreault

Thinking Matters Symposium

The objective of this research project was to create a wearable device that monitors bodily functions for the user to view on their smartphone. Sensor data is processed using the Arduino Nano 33 BLE microcontroller. The sensors used in this project include: proximity, temperature, humidity, heart rate, pressure, and skin impedance. This project takes advantage of the Arduino's Bluetooth low energy (BLE) capabilities so that all the data can be transmitted to a smartphone. This presentation shows the challenges faced during the project and how they were overcome. Some of these challenges include: programming, how heart rate sensors work, and …

Parametric Design Of Flywheel For An Electric Punch Press, Prashant Dhakal

Honors Theses

Almost all designs start with a design requirement. These may be weight, volume, clearance, and even custom calculations. Finding a solution that meets these requirements often involves tedious and time-consuming iterations. Once the solution has been found, the design must be optimized to find the best possible solution. Finding the ‘best possible design’ requires substantial time and effort dedicated to iterating and analyzing the potential model solutions. This is where Parametric design comes into play. The Parametric design aids in specifying the key parameters of the design and makes changes interactively, with the model updating automatically. The purpose of this …

Application Of Atomically Precise Metal Nanoclusters In Electrocatalysis, Zhi-Hua Zhuang, Wei Chen

Journal of Electrochemistry

Metal nanoclusters (M NCs) consist of only several to a few hundred of metal atoms and possess core sizes less than 2 nm. Owing to the quantum size effect, the electronic states of M NCs evolve to discrete levels similar to the molecule energy gaps, other than a continuous density of states to produce plasmon characteristic of bulk metal nanoparticles (M NPs). In comparison with the conventional M NPs, M NCs exhibit dramatically unique electronic and optical properties, such as molecule-like energy gaps, strong photoluminescence and high catalytic properties, which make them promising for potential application in numerous fields, such …

Preparation And Characterization Of “Water-In-Salt” Polymer Electrolyte For Lithium-Ion Batteries, Xu Hou, Xin He, Jie Li

Journal of Electrochemistry

Since the development of wearable and flexible electronic products, the demand of flexible energy storage devices such as batteries and super capacitors is in urgent. To enhance the safety and cycling stability for flexible lithium-ion batteries, “water-in-salt” polymer electrolyte was prepared by introducing 21 mol·kg^-1 LiTFSI electrolyte into cross-linked polyacrylamide (PAM) after freeze-drying. A great amount of holes with the size range of 10 ~ 20 μm can be found on the surface and in the bulk of polyacrylamide, which is benefited from the freeze-drying process and acts as a great support for the electrolyte uptake. The “water-in-salt” polymer …

Lithium Storage Performance Of Hard Carbons Anode Materials Prepared By Different Precursors, Zhen-Lang Liang, Yao Yang, Hao Li, Li-Ying Liu, Zhi-Cong Shi

Journal of Electrochemistry

Hard carbon is one of the most promising anode material for lithium ion batteries (LIBs) owing to its high stability, widespread availability, low-cost, and excellent performance. The electrochemical properties of hard carbon materials depend strongly on the type of precursors. It is, therefore, very important to choose an excellent hard carbon precursor. Polyacrylonitrile, petroleum pitch and peanut shells were used as raw materials to prepare different hard carbon anode materials for LIBs. These hard carbon anode materials were successfully synthesized in two steps. The selected precursor was firstly carbonized at 600℃ for 1 h in argon atmosphere using heating rate …

Oxygen-Exposure Induced Rapid Oxidation Of Spiro-Ometad In Cspbibr2 Perovskite Solar Cells, Wei-Guo Wang, Tian Bai, Gao-Fei Xue, Mei-Dan Ye

Journal of Electrochemistry

2,2′,7,7′-tetrakis(N,N-di-p-methoxyphenyl-amine)-9,9′-spirobifluorene (Spiro-OMeTAD) is the most widely used hole transport material in perovskite solar cells (PSCs). However, its oxidation in the air takes a long time and results in the attack of perovskite by water. In this regard, we performed the oxidation process of Spiro-OMeTAD in oxygen, where perovskite can be protected from water, guaranteeing the integrity of perovskite. It was demonstrated that the champion Spiro-OMeTAD based CsPbIBr₂ PSCs after oxygen oxidation achieved a 7.19% power conversion efficiency (PCE), showing a higher PCE than 6.29% of the champion device oxidized in air. A series of electrochemical characterization methods were applied …

Theoretical Studies Of Metal-N-C For Oxygen Reduction And Hydrogen Evolution Reactions In Acid And Alkaline Solutions, Xue-Ping Qin, Shang-Qian Zhu, Lu-Lu Zhang, Shu-Hui Sun, Min-Hua Shao

Journal of Electrochemistry

Single atom catalysts (SAC) have been regarded as the promising alternatives to platinum group metals due to their low costs and potentially high catalytic activities in various electrocatalytic reactions. The atomic mechanism understanding of activity discrepancy among different metal and nitrogen co-doped carbon-based catalysts is still lacking. Here, non-precious metal and nitrogen co-doped carbons (Me-N-C, Me = Fe and Co) as the model catalysts are investigated by combining experimental and theoretical studies to explore the catalytic activities and corresponding reaction mechanisms toward oxygen reduction reaction (ORR) and hydrogen evolution reaction (HER) at universal pHs. Atomic theoretical simulations suggest that Fe-N-C …

Pd-Based Electrocatalysts For Oxygen Reduction And Ethanol Oxidation Reactions: Some Recent Insights Into Structures And Mechanisms, Zhi-Peng Wu, Chuan-Jian Zhong

Journal of Electrochemistry

The development of efficient electrocatalysts for applications in fuel cells, including proton-exchange membrane fuel cell (PEMFC) and direct ethanol fuel cell (DEFC), has attracted extensive research attention in recent years. Oxygen reduction reaction and ethanol oxidation reaction are two of the key reactions where the design of active, stable and low-cost electrocatalysts is critical for the mass commercializations of PEMFCs and DEFCs. This challenge stems largely from the limited understanding of the catalyst structures and reaction mechanisms. Progress has been made in investigations of electrocatalysts derived from Pd-based alloy nanomaterials both experimentally and computationally. We highlight herein some of the …

Suppressing Bias Stress Degradation In High Performance Solution Processed Organic Transistors Operating In Air, Hamna F. Iqbal, Qianxiang Ai, Karl J. Thorley, Hu Chen, Iain Mcculloch, Chad Risko, John E. Anthony, Oana D. Jurchescu

Chemistry Faculty Publications

Solution processed organic field effect transistors can become ubiquitous in flexible optoelectronics. While progress in material and device design has been astonishing, low environmental and operational stabilities remain longstanding problems obstructing their immediate deployment in real world applications. Here, we introduce a strategy to identify the most probable and severe degradation pathways in organic transistors and then implement a method to eliminate the main sources of instabilities. Real time monitoring of the energetic distribution and transformation of electronic trap states during device operation, in conjunction with simulations, revealed the nature of traps responsible for performance degradation. With this information, we …

Nitro Group Reduction For Use In Organic, Cathodic Materials, Brock G. Goeden

Honors Thesis

The industrial demand for higher capacity, light-weight battery materials has skyrocketed in recent years due to heavy investments in portable electronics, electronic vehicles, and renewable energy sources. However, rechargeable battery technology has seen little improvement since the invention of the Lithium-Ion battery in the 1980s. The low energy density of the traditionally utilized LiCoO2 cathodic material (specific capacity: 272 mAh g-1), has limited its potential to meet these increasing demands. To solve this problem, our research group is investigating new types of lightweight, organic, polymeric materials with conductive backbones as a possible replacement for the cathodic materials in Lithium-Ion batteries. …

Heterogeneously Integrated Photonic Modulators And High-Volume Manufacturing Of Related Technologies, Nicholas Adam Boynton

Electrical and Computer Engineering ETDs

Silicon photonics is an attractive approach to cost-effective integrated optics due to the infrastructure established for silicon CMOS electronics. The material properties of silicon however are not ideal for optical devices. Specifically, silicon lacks the ability to easily produce light-emitting devices due to its indirect bandgap, and has a centro-symmetric crystal structure which does not facilitate the Pockels effect required for linear modulation. Conversely, lithium niobate is an excellent optical material due to its strong Pockels effect but, is a notoriously difficult material to process. One method of simultaneously overcoming the material limitations of silicon and the fabrication limitations of …

On The Dynamic Generation Of Megagauss-Level Magnetic Fields On 100-Ns Timescales To Stabilize And Magnetize Pulsed-Power-Driven Liner Implosions, Gabriel A. Shipley

Electrical and Computer Engineering ETDs

This dissertation presents analysis of experiments and simulations executed to develop the auto-magnetizing liner concept (AutoMag) for use as an alternative premagnetization mechanism for MagLIF. Tests of each stage of AutoMag (magnetization, dielectric breakdown, and implosion) were executed on the Mykonos accelerator and the Z accelerator. Experiments demonstrate strong peak axial magnetic field production (20 – 150 T), dielectric breakdown initiation that depends on global induced electric field across the target, and a level of cylindrical implosion uniformity high enough to be useful for prospective fusion-fuel-filled (auto-magnetized MagLIF) experiments.

This dissertation also presents detailed simulations of the Solid Liner Dynamic …

Design Of Smart Trashcan, Haoran Song

Senior Theses

A smart trashcan has been designed which can bring convenience to people for throwing their garbage away during the COVID-19 pandemic. A prototype is made from cardboard to demonstrate its function. This trashcan can sense people who are coming and leaving, and it can open and close automatically. The trashcan is powered by solar energy. A solar panel is mounted on top of the trashcan supporter. This design is specifically for use in China.

Unsupervised Multivariate Time Series Clustering, Md Monibor Rahman, Lasitha Vidyaratne, Alex Glandon, Khan Iftekharuddin

College of Engineering & Technology (Batten) Posters

Clustering is widely used in unsupervised machine learning to partition a given set of data into non-overlapping groups. Many real-world applications require processing more complex multivariate time series data characterized by more than one dependent variables. A few works in literature reported multivariate classification using Shapelet learning. However, the clustering of multivariate time series signals using Shapelet learning has not explored yet. Shapelet learning is a process of discovering those Shapelets which contain the most informative features of the time series signal. Discovering suitable Shapelets from many candidates Shapelet has been broadly studied for classification and clustering of univariate time …

Data-Limited Domain Adaptation And Transfer Learning For Learning Latent Expression Labels Of Child Facial Expression Images, Megan Witherow, Winston Shields, Manar Samad, Khan Iftekharuddin

College of Engineering & Technology (Batten) Posters

While state-of-the-art deep learning models have demonstrated success in adult facial expression classification by leveraging large, labeled datasets, labeled data for child facial expression classification is limited. Due to differences in facial morphology and development in child and adult faces, deep learning models trained on adult data do not generalize well to child data. Recent deep domain adaptation approaches have improved the generalizability of models trained on a source domain to a target domain with few labeled samples. We propose that incorporating steps of deep transfer learning, e.g. weights initialization from the pre-trained source model and freezing model layers, may …

Twisted Spatiotemporal Optical Vortex Random Fields, Milo W. Hyde Iv

Faculty Publications

We present twisted spatiotemporal optical vortex (STOV) beams, which are partially coherent light sources that possess a coherent optical vortex and a random twist coupling their space and time dimensions. These beams have controllable partial coherence and transverse orbital angular momentum (OAM), which distinguishes them from the more common spatial vortex and twisted beams (known to carry longitudinal OAM) in the literature and should ultimately make them useful in applications such as optical communications and optical tweezing. We present the mathematical analysis of twisted STOV beams, deriving the mutual coherence function and linear and angular momentum densities. We simulate the …

Guest Editorial: Edge Intelligence For Beyond 5g Networks, Yan Zhang, Zhiyong Feng, Hassnaa Moustafa, Feng Ye, Usman Javaid, Chunfen Cui

Electrical and Computer Engineering Faculty Publications

Beyond fifth-generation (B5G) networks, or so-called "6G", is the next-generation wireless communications systems that will radically change how Society evolves. Edge intelligence is emerging as a new concept and has extremely high potential in addressing the new challenges in B5G networks by providing mobile edge computing and edge caching capabilities together with Artificial Intelligence (AI) to the proximity of end users. In edge intelligence empowered B5G networks, edge resources are managed by AI systems for offering powerful computational processing and massive data acquisition locally at edge networks. AI helps to obtain efficient resource scheduling strategies in a complex environment with …